Injection molding is widely used to process short fibre reinforced thermoplastics. The quality and especially the mechanical properties of the resulting part are linked to the mold conception (for example the gate(s) and the venting port(s) locations) and to the processing parameters which will govern fibre orientation distribution. Fibre orientation modelling is based on the well known Folgar and Tucker equation. The models differ one from another by the interaction parameter, the closure approximation and by the coupling with the rheology of the reinforced melt. Quantitative comparison with experiments is very tedious and generally limited to simple part geometries (plaque or disk). As a consequence, in complex geometries, fibre orientation distribution is experimentally checked using several techniques and the resulting an isotropic thermo-mechanical properties are computed using various homogenization theories. In this paper, we propose a first integrated approach of the injection molding of fibre reinforced thermoplastics starting from rheology of the material, orientation equation, interaction parameter and closure approximation. The resulting local fibre orientation distribution is then used in two ways in order to predict the mechanical properties of the part: first, using classical analytical homogenization theories, but based on the computed orientation tensor and not on an experimental one, and then, using numerical homogenization which consists in generating a Representative Elementary, Volume (REV), determining its unidirectional mechanical properties and finally, in computing directly the an isotropic properties of the part.

In this work, we describe a numerical technique to predict fiber orientation during injection moulding of fiber reinforced polymers, and how the resulting part behaves regarding this process induced orientation. The orientation state of a set of fibers is described by a second order tensor. Its evolution is given by the Folgar and Tucker tensorial hyperbolic equation. Even if this equation contains a fourth order term, it may be expressed as a function of the second order tensor using a closure approximation. The resolution of Folgar and Tucker’s equation is carried out by a continuous approach based on the Standard Galerkin method, with stabilisation. The results are compared with experimental orientation measurements on an injected plate. Once the part solidifies it is considered as a biphasic material, composed by the fibers and the polymer matrix, where each phase has a linear elastic behaviour. The fhermo-elastic properties of the composite material are linked to the fiber orientation and the properties of each phase using a homogenisation technique. Finally, to validate the previous study on the prediction of the thermo-elastic properties at the solid state, a three-dimensional industrial case is deeply analysed.

This work’s context is an industrial project aiming the accurate modeling of the injection molding process. 3D numerical simulation of the different stages is considered: during processing, anisotropy of the stress state build up affects its mechanical, optical or dimensional properties, and induces warpage once the part is ejected.A first example of injection molding of reinforced thermoplastics will be treated. In this case, we will consider that during the injection step, an orientation will be induced by the flow. Furthermore, the thermoplastic matrix will pass from the liquid to the solid state, and orientation and stresses will remain frozen. Evolution of orientation or extra stress is computed using the Folgar and Tucker equation, with continuous or discontinuous approximations. Results are obtained in a 3D complex industrial part.

In this paper, we present a strategy of Arabic words recognition by combining two levels which are based on global and analytical approaches according to the topological properties of Arabic handwriting. In the first level (global), we consider the visual indices which can be generated by: diacritics and strokes (denoted tracing) that form the main shapes of the word. Each word is described as a sequence of visual indices which is treated by a “global” classifier based on Hidden Markov Model (HMM). In the second level, the word is segmented into graphemes, then each grapheme is transformed into a HMM observation by a vector quantization phase. An analytical HMM is developed in order to manage the observation sequences. At this level the diacritics are not taken in consideration which allows to reduce the number of estimated character models. Finally we combine the two approaches to decide on the class of an unknown word. In fact, the global model serves as a filter. It produces a set of hypotheses to the analytical model, which in turns, defines and outputs the final decision.

In this paper, we show how planar hidden Markov models (PHMM) can offer great potential to solve difficult Arabic character recognition problems, especially its cursivness. A convenient architecture is defined for printed Arabic sub-words. It yields an easy solution to implement the modeling of the different morphological variations of the Arabic writing, i.e., vertical and variable horizontal linkages. A more flexible architecture, developed for Arabic handwritten words, is under test. The structure proposed presents the aptitude to absorb the variability of the manuscript. Indeed, the experiments have shown promising results and directions for further improvements. In the present paper, we describe both retained architectures, showing the applicability of the PHMM to the Arabic complexities. This is owed precisely to the definition of the PHMM, which permits to follow efficiently the natural variations in bands of the Arabic script.

A perfect segmentation method would be capable to segment words in letters. It would be then possible to define a process on letters. Unfortunately, such a method is almost impossible to obtain due to the nature of handwritten words. To tackle this problem, our approach segments the word into graphemes. We propose in this paper an analytical approach based on the Hidden Markovian Models (HMMs) to manage the defaults of the segmentation module. We also selected an optimal alphabet of graphemes in order to increase the performances of the recognition system. Furthermore, HMMs being developed exploit and model the notion of sub-words that is inherent to Arabic handwriting. An average correction of recognition rate of over 82.5% is obtained (in the first rank) with a lexicon of 232 different Tunisian state names.

A fast recognition method for Arabic handwritten characters is proposed. The recognition problem is divided in several tasks and distributed to appropriate agents. In order to segment Arabic words into graphemes we analyse the upper contour of the connected parts of words. This signal is used to the detection of primary segmentation points PSP. A local analysis gives the decisive segmentation points, DSP. Each word is analysed with its characteristics and the decision is calculated with a maximum likelihood classifier. A second classification is performed with the modelisation by Hidden Markov Models (HMM) applied on the graphemes. The results of the two classifiers are discussed.

We propose a segmentation method and handwritten word coding method by human observation for automatic document processing in Arabic. The system is composed of three levels. The first level deals with the word segmentation into portions of characters called graphemes. The second level analyses these graphemes and codes the word by a sequence of observations similar to human perception. The results of these two levels are used in the recognition level (the third level) which are presented as perspective in this paper

In this paper, we propose two methods of character segmentation for Arabic handwritten characters and cursive Latin characters. Classical horizontal and vertical projections detect the lowercase writing area in lines. The problem of overlapping lower or upper strokes is resolved with a contour-following algorithm which starts in the lowercase writing area and labels the detected contours. In the first method, the junction segments connecting the characters to each other are detected by taking into account the writing line thickness. The second method detects the upper contour of each word. The strokes are detected in order to find primary segmentation points (PSP). These points are analysed with an automaton that considers the shape of the word for the determination of definitive segmentation points (DSP). The two methods are compared and the results are discussed.

In this paper, we propose two methods of character segmentation for Arabic handwritten characters and cursive latin characters. Classical horizontal and vertical projections detect the lowercase writing area in lines. The problem of overlapping lower or upper strokes is resolved with a contour-following algorithm which starts in the lowercase writing area and labels the detected contours. In the first method, the junction segments connecting the characters to each other are detected by taking into account the writing line thickness. The second method detects the upper contour of each word. ne strokes are detected in order to find primary segmentation points (PSP). These points are analysed with an automaton that considers the shape of the word for the determination of definitive segmentation points (DSP). The two methods are compared and the results are discussed.